Food Structure Food Structure

Volume 8 Number 1 Article 5

1989

Heat-Set Gels Based on Oil/Water Emulsions: An Application of Heat-Set Gels Based on Oil/Water Emulsions: An Application of

Whey Protein Functionality Whey Protein Functionality

R. Jost

F. Dannenberg

J. Rosset

Follow this and additional works at: https://digitalcommons.usu.edu/foodmicrostructure

Part of the Food Science Commons

Recommended Citation Recommended Citation Jost, R.; Dannenberg, F.; and Rosset, J. (1989) "Heat-Set Gels Based on Oil/Water Emulsions: An Application of Whey Protein Functionality," Food Structure: Vol. 8 : No. 1 , Article 5. Available at: https://digitalcommons.usu.edu/foodmicrostructure/vol8/iss1/5

This Article is brought to you for free and open access by the Western Dairy Center at DigitalCommons@USU. It has been accepted for inclusion in Food Structure by an authorized administrator of DigitalCommons@USU. For more information, please contact digitalcommons@usu.edu.

FOOD MICROSTRUCTURE, Vol. 8 (1989), pp. 23-28 0730- 5419/89$3 .00+. 00 Scanning Microscopy International, Chicago (AMF O,~lore), IL 60666 USA

HEAT-SET GELS BASED ON OIL,AIATER EMULSIONS: AN APPLICATION OF \JHEY PROTEIN F1JNCTIONALITY

R. Jose*, F . Dannenberg and J. Rosset

Nestle Research Centre, Nestec Ltd. Vers- chez - l es- Blanc

CH-1000 Lausanne 26 (Switzerland)

The microstructure of protein/ lipid gels produced by heat treatment of whey protein sta· bilized oil-in-water(Of\l) emulsions was studied. Scanning e l ectron microscopy as well as trans­mission electron microscopy was performed on glu­taraldehyde/osmium tetrox ide fixed , and critical point-dried samples. Microstructure analysis showed that in the case of homogenized 0,/\.J emul­sions, extensive coat.ing of the fat globule sur­face wic.h coagulated protein led to a "cauli· flower-like " structure. In such gels, uncoated fat globules having smooth surfaces were no t present, This specific microstructure was not obtained wl.th protein/lipid gels in whic h whey vt·ot~in wo s added in the continuous pha!JO nnd mixed with a 0 /U emulsion stabilized by lecithin.

Initial paper received February 14 , 1989 Manuscr ipt received May 10, 1989 Direct inquiries to R. Jos t Telephone number: 41 -2 1 785 81 11

KEY UORpS : Oil/Water emulsions, gelation, homo · genization , lipids, microstructure, protein, scanning elect ron microscopy, transmission elec­tron microscopy, whey protein.

*Address for correspondence:

R. Jost, Nestle Research Cent r e, P.O.Box 3S3, CH-1800 Vevey,

23

Introduction

The u:ost important functional property of whole egg is its ability to form heat-set gels. This "functionality" is exploited to form textural properties in culinary applications of egg. To design a functional egg substitute from whey protein, we developed a process in which whey protein concentrates are combined with selected lipids of vegetable or animal origin. High pressure homogenization o f such compositions leads to stable 0/W emulsions in which whey protein acts as emulsifier mill. at a subsequent process stage, as the gelling agent (Jose ct. a l ., 1986).

A remarkable feature of the gelling 0/W emulsion is the fact that lipide, <llthough having no gelation capacity on their own, contribute to the firmness of the gelled texture, under the condition that they are presen t as a homogeneous population of small droplets. Critical droplet size and size distribution parameters were, e.g., 700 JUll diameter for the mode of t he distribution as an upper limit and a polydispersity index of < 0.35 (Baechler et al., 1986).

Another feature of gels produced from such 0/W emulsions is their smooth texture and lipid· like "mouth-fee l". Microstructure studies of such protein/lipid gels were aimed at finding charac­teristics in their fine structure which might explain textura l properties and elucidate the role of high pressure homogenization ,

Materials and Methods

0/W emulsions were produced from whey pro­tein concentrates (aqueous phase with 5-10 % w;v of \JPC solids at pH ranging from 5. 5 to 8 .0) and from sunflower oil as previously described (Jost et al., 1986). The composition used for micro· structure studies was an emulsion with 15 \ (v/v) of sunflower o i l and 7. 5 \ UPC sol ids , correspon· ding to 6 % of protein. The pH of our emulsions was adjusted to 7.0 by addition of KOH(l N) or NaOH(l N), prior to gelation. Heat·induced gela· tion of the degassed Of'W emulsions was performed in hermetically closed glass beakers (25 ml total volume, 27 mm inner diameter), placed in a metal rack for the incubati on in a water bath at 90•c,

R. Jost, et: al.

for 30 min. Following cooling to room tempera­ture, the gels were aged 24 h at 4•c prlor to further processing.

Measurement of particle size distribution in emulsion was done by dynamic light-scattering, using a Malvern photon correlation spectrometer (Masson and Jost, 1986)

Scanning electron microscopy <SDO Gel slices (1 mm x 2 mm x 6 mm) were fixed

in phosphate (0.1 M, pH 7. 2) buffered 3 \ glutar­aldehyde . Fixation was performed for about 12 h and was followed by post- fixation in 2.5 \osmium tetroxide for 2 h . Glutaraldehyde fixation had been shovn to preserve particularly well the original structure of protein gels ( Heertje and van Kleef. 1986). Critical point-drying in 00:2 following the dehydration in alcohol was pre ­ferred to freeze-drying, thus minimizing the risk of ice crystal formation (Yood~o~ard and Cotterill, 1985).

Particles which had been freshly dry-frac­tured vere mounted on SEM stubs wi th a conductive carbon cement and coated with a 20 nanometer layer of gold in a Polaron sputt:er coating unit. The preparations \Jere observed in a Philips 505 SEM at an accelerating voltage of 30 kV ,

For fixation of the (liquid) 0(\J emulsion, "deep-well" stubs ""ere used. A drop of the emulsion ""as placed in the cavity of the stubs and covered wi th a Nucleopore filter membrane (0.4 J,Jm), and sealed with a me tal ring . At chis scage, the stubs were processed in an analogous manner as the gels.

Transmission electron microscopy <TEM) Small cubes of gel (1 mml) 1oo1ere fixed as

described for SEM. Dehydration ""as accomplished in a graded alcohol series up co 95 ethanol concentration. Infiltration was with mixtures of 95 \ alcohol and ut Yhit:e (2/1 vfv overnight and 1/2 vfv for another 12 h) . Subsequent infiltra­tion of pure ~ Yhite in gelat:in capsules was performed overnight at 4•c, follo""ed by polymeri­zation at sz·c. Thin sections (60 nm) stained with uranyl acetate and lead citrate were examined in a Philips 300 TEH at an acceleration voltage of 80 kV.

Results and Discussion

SEM of an unheated 0/W' emulsion (15 % v/v of dispersed oil and 6 % ""/v of whey protein, pH -7) shows primarily the fat: g lobules in a per­fectly spherical shape. The mean diameter estab­lished by light scattering analysis on the liquid parent. emulsion, roughly 400 run, corresponds well with the projected globule diameters as revealed by SEM (Fig. la). The protein which is essenti­ally present in a soluble state is of rather discrete appearance or not visible at all.

Following heat- induced gelation of the emulsion, SEM sho""s an entirely different struc­ture. Again, the lipid droplet:s are very clearly visible but, in the gel, their surface is rough and structured (Fig. lb) and at a higher magni ­fication, a "cauliflo1oo1er·like" structure is evi-

24

dent: (Fig. lc). The granulated surface structure of the fat: globules is thought to result from extens ive protein coagulati.on on the lipid sur­face. Smooth lipid globules are not seen nor can we distinguish protein coagulum not associated ""ith the lipid globules, at least not: in consi­derable amounts.

il&............. OjiJ emulsion stabilized by whey protein and its heat-set gel. a: SEM of the unheated emulsion . b and c: SEM of the resulting gel.

Yhey protein gels based on OfiJ emulsions

~· SEH micrograph of a whey protein gel produced from an aqueous dispersion of the protein.

It is very instructive to compare the fine s truc ture of the gels based on the emulsion with a whey prot.ein gel produced under the same hea­ting conditions using a mere aqueous dispersion of whey protein concentrate (12 % wjv of total s olids or 9.6 %of protein). Such a protein gel appears in SEM as a dense but porous structure composed of strands o.f protein granules (Fig. 2), The diameter of the protein granules which are roughly spherical, is near 100 run. We can imagine that the same kind of protein granules melted together on the fat surface, leCtds to a micro­s tructure similar to that shown in figs . lb and lc .

We learn from this comparison that: in an OfiJ

emulsion stabilized by whey protein and possibly other proteins, the dispersed lipids when coated with the protein,act as a matrix on which further protein will preferentially coagulate during the he ating step. TEM of the gelled emulsions brings additional strong evidence in favour of a massive interfacial protein coagulation, resulting fro11 the preceding homogenization step during and after which protein "migrated" from the continu­ous phase to the interface.

Thin sections show the protein coating of the lipid globules (Fig. 3a) and the higher magnification clearly shows how gelled protein cross links the coated fat globules (Fig. 3b) , Additional experiments were done to high-light the particularity of the gelled emulsions and the r ole of homogenization in t:he presence of the p ro ~ein for the formation of this structure . It appeared that: the formation of a uniforlll and integrated protein/ lipid microstruct:ure could not be obtained unless homogenization at relatively high pressures (20-35 HPa, 2-5 passages) in the presence of the protein, preceded the gelation seep . Stabilization of the emulsion resulting from protein migration to the interface is necessary to favour protein coagulation at the 1 ipid surface. Varying the proport:ions of lipids and proteins will certainly modify the fine struct:ure of the gel. As we increased the prot:ein content of the emulsion and consequently t:he protein/lipid rat:io, more protein coagulation

25

Ii&...........l . Cel obtained from a whey protein-based OjlJ emulsion. a and b: TEM micrographs of thin sections of the gel.

took place in the continuous phase , while at protein/lipid ratios< 0.5, most of the protein was associated with the fat globules.

A good demonstration that interfacial pro­tein coagulation depends on the presence of the protein during the homogenization step was ob­tained from SEK analysis of gels produced from a ~-stabilized Of\J emulsion, to which whey protein was added. The protein was simply well

R. Jost, et al.

dispersed in the emulsion and the combined ernul· sian subject:ed to heat treatment. Gels obtained 3Ccording to such a procedure were considerably weaker than their counterparts produced with ~ stabilized emulsions. Weak gels vere formed provided that the lecithin concentration was < l 'i., whereas at higher lecithin concentra· tions, the emulsion remained completely liquid after heating. SEM. of lecithin·based gels s how smooth lip id spheres, devoid of adsorbed protein, rn.ndomly distributed in the gel matrix made of whey protein (Fig. 4).

Conclusions

The particular texture and microstructure observed in whey protein/lipid gels reflects the sequence of operations during processing (Fig. 5 ) High pressure homogenization in the presence of protein is . in view of the desired gelation pro?erties , to be preferred to the use of an emulsifier such as lecithin during the homogeni· z.ation step required to produce the parent: Of\J emulsion . Subsequenc heat- induced gelacion then results in a particular "integrated" microstruc· ture characterized by an extensive procein coagu· lation at the surface of the fat globules. Repea­ted homogenization at pressures of 20-40 HPa in · c reases the fat surface available for protein adsorption and coat ing and thus favours the sub­sequent deposition of heat-coagulated protein at the interphase. With an increasing homogeneity of the dispersed lipids and a reduction of the fac globule s i ze, the heat-set gel increases in its strength. Apparently, under these conditions, the fat globules assume the role of filler part.icles as proposed by Dickinson et. al. (1985) who sho\oled this effect. wich gelatin-based 0/lJ emulsions.

Due to the volume fraction which is occupied by the fat globules in emulsion, the protein concentration is also increased in the continuous phase. This explains ""hY in an~. firm gels can be obtained at comparatively lower protein concentrations than in the case of aqueous dispersions of the protein.

In gelled OflJ emulsions, proteins and lipids form an integrated particular microstructure which is macroscopically smooth and gives a pleasant "mouth-feel". This may be favourably explo ited in different food applications.

Acknowled gements

Th P. authors wish to acknowledge excellent technical assistance provided by Claire·lsabelle Heritier.

Baechler R, Fosseux Y, Jost R (1986) Prepa· radon of gelled food products . Eur. Patent EP 0 195 365.

Dickinson E, Stainsby G, Wilson L (1985) An adsorption effect on the gel-strength of dilute gelatin-stabilized oil-in-\olater emulsions. Colloid & Polymer Science 263, 933-934.

Heertje 1, van Kleef FSH (1986) Observations on the microstructure and rheology of ovalbumin gels. Food Microstructure .2_, 91-98.

26

Jost R, Baechler R, Hasson C (1986) Heat gelation of oil-in-water emulsions stabilized by whey protein. J. Food Science ll. 440·4114,

Hasson G, Jost R (1986) A study of oil- in· water emulsions stabilized by ""hey proteins. Colloid and Polymer Science lli. 631·638.

Woodward SA, Cotterill OJ (1985) Preparation of cooked egg white, egg yolk, and whole egg gels for scanning electron microscopy. J. Food Science iQ, 1624- 1628.

Discussion with Reviewers

Sheng· Chin Yang: Can you explain \olhy you use so·c as the emulsification temperature? What if the temperature is lower than so·c? ~: By raising the temperature during homogenization, the reduction of fat globule size is facilitated. However, an upper limit is given by the risk of protein denaturation, and, in t.he case of whey proteins, this limit temperature is 60•c at neutral pH. Homogenization at. so•c is therefore a compromise between a good homogeni­zation effect and the risk of protein denatura­tion. Homogenization at so•c avoids unwanted denaturation at the st:age of the emulsion.

~: Does "'hey protein serve as an ernul· sifier in the unheated emulsion? If so, "'hat happens to the emuls ifier whey protein when t.he system is heated? Is that protein also denatured/ coagulated? Does the heating alter the protein's emulsifying capacity and the emulsion stability? Authors: The whey protein is t h e .2Jll.x. emulsifier present in the system. Following homogenization, a stable emulsion is obtained in which no coa­lescence nor crea111ing is observed, nor does the fat globule size distribut.ion change signifi· cantly over several days. The experimental con· ditions select:-ed for gelling the e111ulsion (pH near 7, temperature 9o•c, holding ti111e 30 min), suggest: that practically lll the protein, adsor­bed or not, is fully denatured during the heat treatment. As a result of heat·denaturat.ion, a rearrangement of protein molecules adsorbed to the fat globule surface, may have occurred. In practice it is no more possible co distinguish between adsorbed and "non-adsorbed" protein once the gel is formed. Following heat-induced gela­tion , the initial emulsion is "frozen" in a gel led structure , It cannot change fo r this rea· son, unless the gel is dehydrated or destroyed by other means. Microscopic analysis of the gels indica ted project:ed fat. globule diameters which are close to the diameters established by light· scattering on the liquid parent emulsion.

Holcomb : lt is mentioned that protein coagu­lation is seen in the outer phase 'Jhen the pro­tei:'! concentrat-ion is increased. At what. protein concentration is such coagulation visible? &U.h2.fi: The decisive factor is the protein/ lipid ratio, not the absolute protein concen­tration. lf this ratio exceeds 0.5, there is an increasing amount of agglomerated protein obser· ved in the space between the protein -covered fat globules ( Fig. 6). This agglomerated protein shows a structure resembling the one seen with protein gels produced from aqueous dispersions.

Whey protein gels based on 0/V emulsions

fi.&......___!, Gel obtained from a lecithin-stabilized enulsion with whey protein used as the gelling ~ent in the continuous phase.

••••• }')t","f~l~t:f","\ •• D HOMOGENIZATION

D HEAT

5

fl&............ , Scheme illustrating steps in the forma­tion of a lipid-rich gel based on a protein­stabilized 0/W emulsion.

27

f.l&..........§ . SEH micrograph of a gel obtained from a whey protein-based 0/fJ emulsion with a protein/ lipid weight ratio of 0.6.

£i.&.........l . "Deep-well• stub used for processing of 1 iqu id 0/fJ emulsion.

Holcomb: Please provide an illustration of the "deep-well" stub. ~: The stub in which we perform the pro­cessing of the liquid emulsion is shown in Flg . 7 . The (inner) diameter of t h e we 11 is 7 mm.

Shcng-Chin Yan g: If the pH and homogenization conditions remain constant, would the heated (9Q•C/30 min) whey protein/lipid emulsion be affected by different heating temperature and time? ~: Concerning the effect of temperature, variations within the range of 85 -9S•c have little effect on the gel-strength, but at tem­peratures below ss·c the gels become markedly weaker . 10•c is the lower limit to obtain gela­tion. Higher temperatures (e . g . by autoclaving the emulsion in hermetically sealed tins), can lead to higher gel-strength . The appropriate holding time at a given temperature depends on the heat transfer in the emulsion. Thus, the 30 minutes specified in this work are adequate for 25 ml of emulsion heated in glass beakers of the size specified in the Materials and Methods scct:ion .

R. Jost, et al.

E N James Jr Based on the levels of -·lact · albumin and ,8-lactoglobulin in whey, and the area of the olljwater interface produced in this example, is there the possib ility of selective adsorption of protein to the surface, thus in· fluencing the composition of the heat-set protein film ? ~: .8·lactoglobulin accoun ted for about 50 \, o-lac talbumin for 20 \ of the total protein in our whey protein concentrate. These proteins, together with mi nor whey proteins, will compete for the fat surfa ce which in our example may be near 500 m2 /100 ml of emulsion (oil volume fraction 0.4, mean fat globule dia.meter 450 nm). \ole didn't study the selectivity of pro­tein adsorption in our system but may mention the findings of Shimizu et al. (1981) Yho showed that at neutral pH, the relative abundance of indivi­dual whey proteins at the fat surface of en O(!J emulsion well reflected their mass proport::ion in t::he Yhey protein concentrate. Thus, fj-lactoglo­bulin was the most abundant protein in the film at neut::ral and alkaline pH, but at acidic pH , a - lactalbumin was more abundant . We deduce that protein adsorption to fat surfaces is strongly dependent on the pH and the isotonic pH of a given protein. Under the conditon of our emul· sions, it is likely that fj-lactoglobulin is a major fraction in the primary protein film .

E N Jaynes Jr What is believed to be the effect of the added lecithi n, and is the order of addition significant? Proteins can displace each other at the interface (Dickinson et al., 1988) and suUlwu dudecyl sulfate can interpenetrate protein films or possibly displace them (Jaynes and Flood, 1985). Is this vhat occurs? ~: Once the emulsion was produced with lecithin instead of protein, whey protein cannot efficient::ly adsorb to the fat globules (see fig. 4). This may be due to charge interact:: ion between the lecithin's polar head and ionic sites on the protein, which will keep the protein off the fat surface. In such a case we may still get a stable emulsion, but no gel or only a very weak one. We cannot fully explain this poor gelation behaviour but assume that lecithin also interacts directly with protein and interfe res with its gelation propert::ies . Lecithin introduced .l..f..kl.r homogeni · zat i o n in the presence o f Yhey protein, likewise i n terfered with gelation . This cou ld indicate t hat in the unh eated emu lsion, l ec i t h i n is able to d isplace prote in f r om the fat surface.

28

~: Is the high pressure (> 150 bar) re­quired in order to obtain a fine and homogenous emulsion or in order to deposit protein at the interface? ~: Although protein migration to the interface takes place at: lower homogenization pressure, high homogenization pressure allows, with a limited number of passages, to achieve a low mean lipid globule size and therefore a large fat surface.

~: Would other proteins show a similar behaviour? ~: We have a limited experience with egg white shoYing that in similar Of'J emulsions high pressure homogenization under adequate tempera­ture and pH condit:ions leads to improved gelation properties.

Additional References

Dickinson E, Rolfe S, Dalgeish DG (1988) Competitive adsorption of osl·casein and ~ ­

c a s ein in oil-in-water emulsions . Food Hydro­c olloids 2, 397-405.

Jayn;s EN, Flood MA (1985) Protein fllms at oil-water interfaces: interfacial tension measu­rements by the static drop method. J. Disper. Sci. Tech . .§., 55-68.

Shimizu M, Kamiya T, Yamauchi K (1981) The adsorption of whey proteins on the surface of emulsified fat. Agric. Biol. Chem. !Q., 2491-2496.